Assistant for Large-Scale Machine Learning Experiments

2. The computer implemented method of claim 1, further comprising receiving a project configuration including a range of machine learning algorithms, a range of hyperparameters for the machine learning algorithms, and evaluation metrics.

3. The computer implemented method of claim 1, further comprising providing the consolidated and clustered experimental results to an interactive dashboard.

4. The computer implemented method of claim 3, further comprising displaying a comparison or trend analysis across the experimental jobs over time.

5. The computer implemented method of claim 1, further comprising generating the plurality of data set drift splits to provide a specific stress test of a machine learning model to determine whether a domain shift with respect to a machine learning model prediction is present.

6. The computer implemented method of claim 5, further comprising generating a pre-deployment recommendation according to results of the specific stress test.

7. The computer implemented method of claim 1, wherein one or more of the plurality of data set drift splits are obtained by splitting the dataset according to one or more labelled features of the one or more input datasets.

8. The computer implemented method of claim 1, wherein one or more of the plurality of data set drift splits are obtained by splitting the dataset either by removing data nearest a random data point in the one or more input datasets or by removing data farthest from the random data point in the one or more input datasets.

10. The system of claim 9, further comprising the interactive dashboard configured to provide the consolidated, processed and clustered experimental results to a user.

11. The system of claim 9, wherein the one or more data set drift splits are generated by the drift split creation engine as a specific stress test of a machine learning model to determine whether a domain shift with respect to a machine learning model prediction is present.